SPECIAL ARTICLE
E-Governance and Corruption in the States
Can Technology Serve the Aam Aadmi?
Jennifer Bussell
A comparative evaluation of one-stop, computerised
citizen service centres in various Indian states has been
carried out in this paper to assess their efficacy. It is
found that the outcomes of policies related to
e-governance in India are not correlated to conventional
variables such as economic development. Instead the
extent to which political parties in power expect such
policies to affect their current and future electoral
statuses affects implementation.
Jennifer Bussell (jennifer.bussell@mail.utexas.edu) is with the Lyndon B
Johnson School of Public Affairs, University of Texas at Austin.
Economic & Political Weekly EPW June 23, 2012 vol xlvII no 25
T
he potential to use information and communication
technologies to improve public service delivery has
been a key topic of policy interest in both the Indian
states and the centre over the last decade. The central govern-
ment’s National e-Governance Plan (NeGP), established in
2006, set out an impressive agenda for developing e-Govern-
ment services, networking infrastructure, state data centres,
and village-level centres for delivering core services to citizens
(Government of India 2006a). This programme built on earlier
efforts by a wide range of states to offer computerised services
such as ration cards, income certificates, building licences,
land records, and income tax payments. From eSeva in Andhra
Pradesh, Nemmadi in Karnataka, and Friends in Kerala to
e-Mitra in Rajasthan, Civic Centres in Gujarat, and Sugam in
Himachal Pradesh, state governments have been leaders in
computerised service centres.
Yet, as the central and state governments continue to imple-
ment the NeGP across the country, we have only a minimal un-
derstanding of how states approached the problem of compu-
terised service delivery in the years leading up to the national
plan, what types of services were made available to citizens,
what obstacles were faced by politicians and bureaucrats in
the implementation of these initiatives, and the extent to
which these programmes are likely to benefit citizens. Ana-
lysts have evaluated individual computerisation initiatives
(see, inter alia, Ahuja and Singh 2006; Bhatnagar and Singh
2009; Caseley 2004; Kuriyan and Ray 2009), but no compre-
hensive comparison of state activities has evaluated compu-
terisation policy outcomes. Did all states implement computer-
ised service centres in a similar manner? Did all states deliver
similar types of services to their citizens through these cen-
tres? If not, what explains the variation in state government
choices over e-Government programmes?
This article addresses these questions on the basis of a com-
prehensive study of state e-Government initiatives during the
period 1999-2009.1 In particular, I evaluate state efforts to in-
troduce one-stop service centres, similar to the common service
centres promoted in the NeGP. This analysis offers two impor-
tant findings. First, while nearly all major Indian states imple-
mented some type of computerised service centre programme
during this period, the programmes themselves differed sig-
nificantly in terms of the number and type of services made
available to citizens (Figure 1, p 78). This substantial diversity
in strategies taken by state governments highlights the dra-
matic ways in which the potential benefits of reformed service
77
SPECIAL ARTICLE
delivery are limited to those citizens in only a few states and
only certain populations within the states.
Second, this variation in policy design and implementation
cannot be explained based on variations in economic or social
conditions, levels of pre-existing technology infrastructure, or
the electoral conditions in a given state. Instead, in order to
explain policy outcomes, it is necessary to understand the
expected effect of these policies on the economic resources of
incumbent politicians and in particular politicians’ expectations
about the threat of more transparent service delivery to estab-
lished sources of corrupt income. While there was considerable
initial enthusiasm to use new technologies, the actual benefits
offered to citizens are constrained in many cases by persistent
efforts to retain access to a rich source of corruption: the bribes
citizens pay to get the services they are promised by the state.2
Understanding Variations
Erstwhile Chief Minister Chandrababu Naidu in Andhra
Pradesh gained significant media attention for his efforts to
computerise government services – everything from electric-
ity bills and property tax payments to caste and income certifi-
cates – and to offer them in one location. The potential
Figure 1: Maximum Number of Available Services Per Centre
in the States (2006)3
0 5 10 15 20 25 30 35 40
Uttarakhand
Delhi
Orissa
Haryana
Tamil Nadu
West Bengal
Kerala
Rajasthan
Uttar Pradesh
Maharashtra
Punjab
Gujarat
Karnataka
Himachal Pradesh
Chhattisgarh
Andhra Pradesh
improvements to the quality of service delivery seemed vast,
and chief ministers from across the country sent officials to
learn about the “eSeva” model.4 Because citizens have often
faced difficult circumstances when attempting to access public
services, it is not surprising that many state leaders saw this as
an opportunity to increase their standing in the eyes of citi-
zens by improving service delivery.5
But if politicians hope to use higher quality, technology-
enabled services to improve their chances of re-election, then
why did state governments differ so dramatically in their in-
troduction of service centres? One answer might be that
poorer states did not have the economic resources to purchase
78
technology to automate processes and establish service centres.
However, the cost of opening a single centre is relatively low
(Toyama et al 2004) and the cost of basic software and technol-
ogy infrastructure has decreased over the last 15 years. A service
centre programme was within the economic reach of all states
during this period and, as shown below, economic conditions are
not correlated with any major characteristics of state policies.
Alternatively, general political conditions might influence
the likelihood that politicians would want to invest in a pro-
gramme to improve service delivery. Those politicians who
face the greatest electoral threat from their opponents might
be more enthusiastic about introducing a new type of public
good, in the form of improved service delivery, to increase their
chances of retaining power. However, because politicians in
nearly every Indian state face the threat of anti-incumbency
bias (Uppal 2009), the pressure to perform is in important
ways consistent across all states. As a result, it seems instead
that all politicians would have an interest in improving serv-
ices, and it is less probable that small electoral differences had
a strong effect on the character of reform. Other characteristics
of electoral competition such as the characteristics of political
constituencies and vote blocs may instead play an important role
in the implementation of policies, which are considered below.
Contrary to these arguments, it is posited that the primary
45
cause of diverse computerised service centre policies is varia-
tion in the extent to which incumbent politicians expect these
centres to affect their economic resources. As in other con-
texts, politicians weigh the expected costs and benefits of im-
proving public services. Yet, the costs to Indian politicians of
digitising public service delivery arise because new techno-
logies can streamline government processes, and so can also
limit the ability of officials to generate and appropriate bribes
from inefficiencies in service delivery. Where bribes are avail-
able in public service delivery – for example, when citizens
find it necessary to pay “speed money” to access services – pol-
iticians are less likely to be supportive of policies to increase
transparency in administration.
In the Indian political system, politicians have multiple possi-
ble sources of corrupt income, including bribes paid by citizens
for services, kickbacks on contracts with the private sector, and
funds skimmed from official state development programmes
(Bardhan 2005; Davis 2004; Frankel 2005; Jenkins 2006; Old-
enburg 1987; Wade 1985). While recent media attention has fo-
cused on major cases of grand corruption, such as in the Com-
monwealth Games (NDTV 2010a), the Maharashtra housing
scandal (Financial Express 2010), illegal export of mineral re-
sources in Karnataka (Pandey 2010), and 2G spectrum licensing
(NDTV 2010b), it is corruption in the day-to-day activities of the
state in interaction with the public that amounts to a Rs 21,000
crore (Transparency International India and CMS 2005) market
in petty corruption that directly affects Indian citizens.
Whether the source of corrupt income is corporate kickbacks
or bribes from individuals, politicians do not extract rents
through the machinery of the state solely for personal pecuni-
ary gain, but often to enhance their hold on political power
(Frankel 2005; Iype 2004; Wade 1985). Access to bribes can
June 23, 2012 vol xlvII no 25 EPW Economic & Political Weekly

serve as an important electoral resource for incumbents.
Because there is no explicit funding mechanism for Indian state
assembly elections, candidates often fund their campaigns
without guaranteed support from the party (Wade 1985).
While the Election Commission of India (ECI) limits candidate
expenditure in state elections (ECI 2007), spending by political
parties and supporters of candidates is not limited, thereby cre-
ating a loophole that indirectly allows for higher spending by
the candidate, who can attribute expenditures above the limit
to the party or friends (Iype 2004). As one former candidate for
a Lok Sabha seat in Kanpur noted, “I spent within the limit of
Rs 1.5 lakh. But my friends and party put in Rs 20-25 lakh”
(Jayant Malhoutra, as quoted in Rekhi and Shekhar 1996).
The potential for somewhat unlimited financial support has
pushed up campaign spending. For example, in Karnataka’s
2008 assembly elections, one successful candidate stated that,
“the quantum of money being used in elections has been
increasing despite the restrictions. The ECI can’t contain it”
(Kumar, as quoted in Sharma 2008). Analysts estimate that
approximately Rs 4,000 crore was spent across Karnataka’s
224 constituencies (Sharma 2008). This increasing economic
demand of campaigns places pressure on politicians to find
lucrative sources of funding.
While politicians may not take bribes directly from citizens
in order to accumulate resources, they can place pressure on
bureaucrats to gain access to these funds. Because politicians
control the movement of officials between positions within the
bureaucracy, and because these postings are typically not reg-
ulated in a formal manner, the potential arises for manipula-
tion of bureaucratic positions and appointments based not on
policy requirements but on political connections (Bardhan
1997; Vithal 1997). A market for job postings can then emerge
in which bureaucrats are incentivised to pay for job transfers,
leading to an increased demand from politicians for illicit
funds (de Zwart 1994; Wade 1985).
Where bureaucrats also have discretionary power over the
provision of public services to citizens, as is often the case,
bribes from citizens offer a source for acquiring the funds
expected by politicians. Bureaucrats can channel a share of
bribe income to politicians, either in hopes of a transfer or sim-
ply to satisfy a political boss’ demands (Bardhan 2005; Davis
2004; de Zwart 1994; Wade 1985). Politicians then have access
to additional funds with which to finance future elections.
Petty Corruption
As a result of these institutional dynamics, if greater trans-
parency from computerisation plausibly disrupts politicians’
established sources of bribes, then this is a threat to their
ability to be re-elected. Politicians in areas with lower pre-
existing levels of petty corruption should anticipate that re-
forms will only minimally affect their access to income, as the
availability of bribes is low from the outset. However, where
there are higher levels of petty corruption, the opposite is the
case. Political incumbents who are more dependent on corrup-
tion to run their re-election campaigns will perceive improved
service delivery as a threat to their illicit income. Thus, politicians
Economic & Political Weekly EPW June 23, 2012 vol xlvII no 25
SPECIAL ARTICLE
in areas with high petty corruption are less likely than their
peers in lower corruption regions to support the initial intro-
duction of reforms.
While all Indian states display some petty corruption, levels
differ quite dramatically across the states. Figure 2 shows a
measure of petty corruption calculated based on a survey con-
ducted by Transparency International India and the Centre for
Media Studies in which a random sample of Indian citizens
was asked about their experiences paying bribes in 11 govern-
ment departments (Transparency International India and CMS
2005). This measure is based on an index of corruption in all
departments. The survey found substantial variation in the
level of petty corruption across the Indian states: bureaucrats
typically demanded bribes from fewer than 20% of citizens in
states such as Kerala and Gujarat, versus more than 50%
of citizens in Uttar Pradesh or Rajasthan (ibid). As a result, if
the level of petty corruption is linked to computerisation
policy choices, we should be able to evaluate this relationship
across the states.
Figure 2: Petty Corruption in the States
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Kerala
Himachal Pradesh
Gujarat
Andhra Pradesh
Maharashtra
Chhattisgarh
Punjab
West Bengal
Orissa
Uttarakhand
Uttar Pradesh
Delhi
Tamil Nadu
Haryana
Jharkhand
Rajasthan
Karnataka
Madhya Pradesh
Bihar
The same motivations are likely to shape many elements of
policy implementation. Politicians accustomed to higher levels
of petty corruption, if they do implement reforms, should be
less willing to implement a wide range of technology-enabled
public services, so as to minimise the overall threat to incom-
ing bribes. In particular, leaders should resist inclusion in com-
puterised service centres of those services that offer the great-
est potential for bribes, either due to their high demand by citi-
zens, their typical value in terms of bribes, or both. These poli-
ticians are also more likely than their peers to resist the com-
prehensive automation of the service-delivery process, so as
to retain non-computerised steps of service delivery that may
offer opportunities for extraction of bribes.
The demand for economic resources can be magnified in
states led by a coalition government. In single-party systems,
party leaders, in particular the chief minister, can implement
reforms according to expectations about the overall costs and
benefits of reform to the ruling party. In coalition-led states,
79
SPECIAL ARTICLE

however, the formation of the coalition may be based on ex- months since implementation, do not display statistically signi-
pectations about ministerial posts for party members and the ficant relationships with the quantity of services made available.
opportunities for bribe-taking associated with those posts. Perhaps more important than the number of services made
Any threat to the expected economic resources of participat- available is the type of services offered by state governments
ing in a coalition will change the incentives of coalition part- in computerised centres. If politicians in higher corruption
ners to participate in government. When coalition members states are concerned with the threat of computerised service
have enough seats to destabilise the coalition, they can use delivery to bribe-taking, then we would expect those services
this power to resist policy directives, by threatening defection with high potential for corruption not to be included in
from the coalition, and so can influence policy outcomes. computerised centres.
With these potential cost considerations in mind, politicians This is indeed the case, both for services that have the poten-
are also likely to attempt to maximise the electoral benefits of tial to deliver a high magnitude of income from bribes, based
reform by targeting improvements in service delivery to their on the demand for the services and the average bribe paid
preferred constituents. Administrative reform in this context (“high corruption potential” services), and for those services
does not merely provide a “public good” enjoyed by all citizens; with high frequency of bribe payments (“high bribe” services).8
instead, the benefits of digitisation can be highly targeted. By In a regression of high corruption potential services on the state
making available services that will benefit particular groups – level of corruption, using a Poisson model to account for the
such as building licences for business elites or welfare benefits for fact that the dependent variable is a count variable, I find a
those below the poverty line – policymakers can increase the strong and statistically significant relationship. A move from
chances that factions needed for their political survival are the below-average corruption states to above-average corruption
beneficiaries of computerised centres. At the same time, politi- states reduces the likelihood of providing a service with high
cians must balance their efforts to provide targeted goods with
Table 1: Provision of High Corruption Potential Services*
their desire to minimise the threat to existing rents, resulting IV Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
in a trade-off between those services with the highest potential Above average corruption .38** .38** .36** .38** .41* .38** .29***
for acquiring bribes and those offering the greatest potential (-3.30) (-3.32) (-3.43) (-3.30) (-2.50) (-3.27) (-3.91)
benefit to key constituents. As shown below, these political con- Coalition 1.13
siderations strongly shape the particular services that are selected (.42)
Electoral competition (ENPS) .88
for inclusion in one-stop, computerised service centres. (-.98)
SDP per capita .97
Analysing Policy Outcomes (-.04)
Months since initiation (12/2006) 1.00
The effects of threats to corrupt rents from improvements in (.40)
service delivery become evident in an analysis of policy char- Human development .47
acteristics. Of the 20 major Indian states,6 16 states imple- (-1.12)
mented computerised service centres of some type during the Historic land tenure .98
(-.04)
period 1999-2006. However, politicians in those states with N 16 16 16 16 16 16 13
higher levels of petty corruption7 were more likely to delay the Prob > Chi2 .00 .00 .00 .00 .00 .00 .00
implementation of these policies, or not implement them at all, *Poisson models with incidence rate ratios reported and z-statistic in parentheses.
*p < .05 **p < .01 *** p < .001.
than leaders of less corrupt states (Bussell 2010). States led by Sources: Election Commission of India, Reserve Bank of India, Indian 2001 Human
coalition governments also implemented service centre poli- Development Report, Banerjee and Iyer (2005), and author’s calculations.
cies later than their single-party-led peers, all else equal. Table 2: Provision of High Bribe Services*
Other characteristics, such as state domestic product per cap- IV Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7
ita, pre-existing technology infrastructure, and electoral com- Above average corruption .37*** .37*** .36*** .37*** .34*** .38*** .31***
petition, showed no strong relationships with the timing of (-4.09) (-4.08) (-4.21) (-4.10) (-3.76) (-4.07) (-4.49)
policy implementation (ibid). Coalition .99
(-.05)
The implementation of policies can tell us even more about Electoral competition (ENPS) .91
the quality of reforms. An initial measure of implementation (-.95)
quality is the number of services made available in state serv- SDP per capita 1.28
ice centres. For example, some states with service centres pro- (.41)
Months since initiation (12/2006) .99
vide only a few services, implying that citizens must still go to (-.62)
myriad government offices to acquire any other public services. Human development .56
Generally speaking, when governments make more services (-1.06)
available in one-stop centres, the benefits to a wide range of Historic land tenure 1.29
(.67)
citizens should increase. N 16 16 16 16 16 16 13
Here we see that those states with higher levels of corruption Prob > Chi2 .00 .00 .00 .00 .00 .00 .00
tend to make fewer services available to citizens in computerised * Poisson models with incidence rate ratios reported and z-statistic in parentheses.
*p < .05 **p < .01 *** p < .001.
service centres (Bussell 2010). In contrast, measures of alterna- Sources: Election Commission of India, Reserve Bank of India, Indian 2001 Human
tive explanations, such as state domestic product per capita and Development Report, Banerjee and Iyer (2005), and author’s calculations.

80 June 23, 2012 vol xlvII no 25 EPW Economic & Political Weekly

potential for corruption by 62%. These findings hold in multi-
variate models, in which no measures of alternative explanations
showed statistically significant relationships with high
corruption potential services (Table 1, p 80). Among those
high bribe services for which more than 25% of the popula-
tion surveyed reported paying a bribe (Transparency Interna-
tional India and CMS 2005), the level of corruption is also
strongly related to the likelihood of provision. Above-average
corruption states are 63% less likely to provide a high-bribe
service than their lower corruption peers. These findings hold
in multivariate models testing for the effects of alternative
explanations (Table 2, p 80).
Figure 3: Percentage of Services in Centres Targeted to Different
Socio-economic Groups in the Indian States9
0 10 20 30 40 50 60
Uttarakhand
Andhra Pradesh
Karnataka
Rajasthan
Kerala
Tamil Nadu
Poor
Uttar Pradesh Middle/upper
All citizens
Chhattisgarh
Delhi
Himachal Pradesh
West Bengal
Gujarat
Maharashtra
Punjab
Haryana
Orissa
Services also provide differing benefits to citizens across socio-
economic categories. For example, ration cards and welfare
schemes are more likely to benefit poorer citizens, while
wealthier individuals are likely to utilise passport applications or
business licences. Policy choices over whether or not to offer spe-
cific services in computerised centres will affect the benefits for
any given individual. Figure 3 shows the percentage of available
services in each state targeting different socio-economic groups
and highlights the diversity of service provision.
Table 3: State Cases
Lower Petty Corruption
Single party government Chhattisgarh
Coalition government Kerala
These cross-state analyses provide a general picture of the
primary relationship between pre-existing petty corruption and
the nature of service reforms, as well as the secondary effects of
electoral incentives to target service provision. However, these
dynamics, and in particular the effects of coalition government,
are best illustrated through evaluations of individual states.
Here I consider four states representing variation in levels of
corruption and the character of the ruling government. In do-
ing so, I am able to provide insights into the effects of both
petty corruption and coalitions on the emergence and imple-
mentation of computerised service centres, as well as the ways
Economic & Political Weekly EPW June 23, 2012 vol xlvII no 25
SPECIAL ARTICLE
in which ruling government support bases play an important
role in choices over service provision.
Chhattisgarh
Despite the diverse set of issues facing Chhattisgarh, including
below average income per capita and the Naxalite movement,
the state implemented one of the most comprehensive service
centre reforms in India. During the initial Congress government,
the information technology (IT) secretary, Sunil Kumar, and an-
other IT department official, Amit Aggarwal, proposed opening
citizen service centres for acquiring government certificates,
similar to eSeva in Andhra Pradesh. The goal of the project was
to reduce the burden on citizens of accessing services, because,
as one bureaucrat put it, “people had to go to lots of different
70 80 90
offices and pay lots of bribes in order to get their services from
government”.10 The initial system put in place, called Chhattis-
garh online information for citizen empowerment (Choice), was
a relatively simple automation of existing service delivery pro-
cesses that provided services in a one-stop shop.
Shortly after implementation, however, the Bharatiya Janata
Party (BJP) took power in the state and Raman Singh was chosen
to be chief minister. It was unclear whether Singh would con-
tinue to support the Choice centre initiative. The chief minister
set out a strategic agenda of good governance for the state and
according to people associated with service reforms, Singh saw
the use of IT to reform service delivery as a key component of
delivering governance reforms to citizens.11 Despite the fact
that, “when the second state government came in they had ap-
prehensions regarding continuing a project that had been
started by the previous government” (ibid), Singh did not close
down the Choice centres. Instead, he became a supporter of
computerisation. When asked explicitly if the chief minister be-
lieved that the overall electoral benefits were greater than any
potential electoral costs from transparency in service delivery, a
top official in the information technology department replied,
yes, the chief minister sees the benefits as greater than the costs...
[because] the CM...is focused on good governance, on accountability,
transparency, and responsiveness…he does not even question the
political costs (ibid).
This support from the chief minister contributed to the state’s
ability to offer more services in its Choice centres than nearly any
other state. This includes high corruption potential and high
Higher Petty Corruption bribe services such as ration cards and land records. The state
Tamil Nadu also utilises private entrepreneurs to run Choice centres, a deci-
Uttar Pradesh
sion state representatives say is partially based on a desire to fur-
ther reduce corruption in the state. As one official noted,
this does not mean that citizens should not be able to access their govern-
ment, but rather that they should not have to interface with a government
officer for every service. This can only help with reducing corruption.
The CM is very clear on this, which sends a message to everyone (ibid).
At the same time, citizen service centres provided the ruling
government with an opportunity to improve services for spe-
cific portions of the population. In contrast with the services
made available under the Congress government, which re-
flected a strong preference for poorer constituencies, such as
caste and income certificates, the services made available by
81
SPECIAL ARTICLE
the BJP reflect quite different strategic priorities. In particular,
the BJP shaped computerised service delivery to target its own
core constituencies.12 Of the additional services made availa-
ble, 42% benefit middle and upper class income groups (in-
cluding building plan approval, passport applications, and
driving licences) while only 16% target lower classes (includ-
ing ration cards and information on government schemes) and
18% benefit all groups (e g, electoral role status).
The BJP government also focused on urban areas of the state,
where it has strong support. During the period of study, Choice
centres had been opened in only six of the state’s 26 districts,
those with urban municipalities.13 While the state is now imple-
menting the national government’s common service centre
programme in rural areas, and so should eventually have
government service centres in all parts of the state, there was
no plan for rural centres prior to the initiation of the national
government initiative.
Overall, Chhattisgarh’s technology-enabled service reforms are
some of the most robust in India. The programme is limited by its
presence only in urban areas, but the high quantity of services
made available reflects the strategic preferences of a chief min-
ister with minimal need to maintain corrupt income from petty
corruption and an electoral strategy that is aligned with the gen-
eral goals of service reform. At the same time, the selection of
specific services to be included in service centres also highlights
the opportunity that these reforms offer ruling politicians to direct
the benefits of computerisation to their preferred constituents.
Tamil Nadu
Now consider an above-average corruption state, Tamil Nadu,
which was also ruled during this period by single party gov-
ernments. The first service centre initiative in the state, the
Sustainable Access in Rural India (SARI) programme, was
launched in 2001 by a coalition of private actors and eventu-
ally implemented 51 “kiosks” in the Melur taluk of Madurai
district (Srinivasan 2005). These centres provided both basic
government documents, such as birth certificates, as well as
computer education, email, and other private services.
In 2002, the state government, ruled at the time by the All
India Anna Dravida Munnetra Kazhagam (AIADMK), decided
to expand the initiative as an element of state policy. In doing
so, the government also changed the name to Rural Access
to Services in India (RASI), which some observers felt was in-
tended to ensure that the current government received credit
for the project.14 Centres were planned for all areas of the state
and by 2006 more than 2,000 had been implemented.15
The allocation of centres in the state, however, reflects the
specific preferences of the ruling party. While results of elec-
tions during the period show that the AIADMK and its main
opposition the Dravida Munnetra Kazhagam (DMK) had gener-
ally consistent support across economic groups (Prasad et al
2004), votes for the parties were divided across geographic
regions with the AIADMK typically doing better in rural areas
(Thirunavukkarasu 2001). The AIADMK’s RASI programme
provided services that were focused on the needs of rural citizens,
with many non-governmental services such as telemedicine
82
and veterinar y advice provided at the centres.16 No similar ini-
tiative was launched in urban areas and it seems clear that the
ruling party politicians used this technology-enabled service
centre initiative to appeal to their rural base.
The number of centres introduced in Tamil Nadu also belies
the relevance of RASI to citizens. While many citizens may
have found a RASI centre near to where they lived, very few
government services were available in the centres. Only some
information on government schemes, in addition to bill pay-
ments and basic certificates, was provided.
This failure to offer comprehensive government services
through the RASI centres is somewhat surprising, given analyses
in recent years that highlight Tamil Nadu as a technologically
advanced state (Government of India 2006b). In contrast, a rep-
resentative of the state information technology department
posited that RASI had “failed” because the state had not been
ready at the time to provide technology-enabled services.17
One explanation for the discrepancy between the views of
analysts and state officials is that while the state has imple-
mented a number of e-Governance initiatives across multiple
departments, these projects have rarely been fully completed
or maintained.18 Computerisation of some services, such as
land records, was initiated, but these projects were rarely fol-
lowed through to full computerisation, making it difficult to
offer the services in RASI centres. Politicians seem to have
made no effort to pressure departments to include their serv-
ices in service centres, even in those cases where initial com-
puterisation of government processes was already initiated.
Politicians may have promoted technolog y initiatives to the
public and analysts, but they felt little need to risk any possible
income flows through comprehensive reforms.
There is evidence for this same trend in the RASI centres
themselves. An assessment in early 2008 found that the majority
of the centres had stopped operating. While the public had
responded positively to the centres, state representatives did not
provide the necessary support to ensure their success. “The set-
ting up of RASI Centres…virtually eliminated harassment and
corruption in government offices as it avoided the people-staff
interface” (Manikandan 2008). As a result, “the e-Governance
project became a hit among people, much to the annoyance of
government staff, elected representatives and political big-
wigs” (ibid). Government officials failed to support continued
operation of the centres, leading to a reduced number of appli-
cations from citizens, forcing most centres to shut down.
Overall, the introduction of computerised service centres in
Tamil Nadu failed to live up to the expectations of both analysts
and citizens in terms of improving service delivery through com-
puterised, village-level access points. Despite the government’s
perceived interest in utilising information technology to improve
service delivery, the largely failed implementation of initiatives
highlights underlying interests in support of the status quo.
Kerala
Kerala has received much positive media attention as a result
of its initiatives in the areas of IT and governance, but the ef-
fects of the state’s one-stop centre initiatives on the quality of
June 23, 2012 vol xlvII no 25 EPW Economic & Political Weekly

public service provision to citizens are minimal. The services
offered in Kerala’s two main service centre initiatives, FRIENDS
(Fast, Reliable, Instant, Efficient, Network for Disbursement of
Services) and Akshaya, pale in comparison to their counter-
parts in many other states. Through FRIENDS centres, only 10
government services have been offered, and just three were
available in Akshaya centres.
The limited introduction of computerised services in Kerala
is initially surprising, given low levels of petty corruption rela-
tive to other states. However, the explanation for Kerala’s pol-
icy outcomes lies in the relationship between corruption and
the dynamics of coalition politics. Kerala’s government is re-
nowned for the persistence of a bi-polar coalition party system
(Heller 2000; Kumar 2004; Nirmala 2006).19 Between 1996
and 2001, the Left Democratic Front (LDF), led by the Commu-
nist Party of India (Marxist) (CPI(M)), controlled the Kerala
state assembly. This government’s eventual emphasis on IT
was unexpected. During the mid-1990s, leftist parties in the
state opposed computerisation of the bureaucracy, expecting it
would lead to job losses in the state (Mathew 1998). However,
in 1998 the government introduced its first IT policy, which
included an e-Governance mandate, “to identify the govern-
ment departments that have a high level of public interface
and to introduce Information Technology to ensure effective
delivery of citizen services” (Madon and Kiran 2002: 2). Not
only did the LDF want to promote the growth of an IT sector in
the state, it also wanted to ensure that these technologies were
used in ways that benefited the “masses”.
The Kerala State Information Technology Mission (IT Mis-
sion) was tasked with implementing government technology
programmes. The first major services initiative was FRIENDS, a
set of computerised centres implemented in district headquar-
ters during 2000-01. These were “single window” service cen-
tres where “citizens have the opportunity to pay all taxes and
other dues to the Government under one roof at no extra cost”
(Kerala State IT Mission 2009). Thus, like most service centre
reforms, FRIENDS centres were intended to reduce the time
and cost to citizens of interacting with multiple different
government departments.
The potential electoral benefits of this improved service
delivery seemed obvious to politicians. As one analysis notes,
“The project was launched at Thiruvananthpuram [the capital
of Kerala] in 2000, in a rather hasty manner, just before the
Assembly Elections” (APDIP 2003). Then, when the LDF gov-
ernment lost control of the assembly in those elections, the in-
coming United Democratic Front (UDF) coalition, led by the
Indian National Congress, further implemented the initiative
in other district headquarters.
Political support for the FRIENDS centres, however, had its
limits, as is clear when we consider the services made availa-
ble. First, despite Kerala’s relatively low level of petty corrup-
tion, FRIENDS emphasises bill payments, which typically in-
volve the lowest levels of corruption.20 Even where payments
can be made for certain types of certificates, citizens at least
initially had to go to the relevant department to pick up the
certificate (Sreekumar 2002). Second, the selection of specific
Economic & Political Weekly EPW June 23, 2012 vol xlvII no 25
SPECIAL ARTICLE
services reflects coalition dynamics within both the LDF gov-
ernment and its successor UDF government. As one analyst
argues, the difficulty associated with negotiations between
departments over provision of services through service centres
“was particularly evident in Kerala, which is ruled by a coali-
tion government, with different political parties in charge of
different departments” (Kiran 2002).
The importance of coalition politics implies that the alloca-
tion of ministerial posts should be related to the selection of
services provided in technology-enabled centres. The CPI(M)
led the LDF coalition with its partner the Communist Party of
India (CPI) holding the next largest number of seats in the state
assembly. The coalition in total held 76 seats, five more than the
necessary majority of 71. When FRIENDS was launched, the cen-
tres offered six main services, bill payments for electricity, tele-
phone, water, property tax, driving licences, and vehicle regis-
tration. Of these services, two of the responsible departments
were overseen by supporting coalition members, the water de-
partment and transport, held by the Revolutionary Socialist
Party (RSP) and an independent member of the legislature, re-
spectively. However, because the RSP held only five assembly
seats and the independent held only one, neither minister had
the power to bring down the coalition government on their
own: even defection by all five RSP legislators would leave the
coalition with sufficient seats to maintain its majority. The
remaining services were delivered by departments overseen by
CPI(M) ministers. No services controlled by the CPI – the only
party with enough assembly seats to threaten coalition stability
– were made available. This is despite the fact that the CPI con-
trolled 15 services that have been offered in other states.
During this same period, a second major service centre ini-
tiative, Akshaya, was launched to provide more affordable ac-
cess to information technologies in rural areas of the state,
through the implementation of village level computer centres.
The government’s primary agenda for Akshaya was “to achieve
the twin goals of social development through access to com-
puters in rural areas and financial viability through market-
driven entrepreneurship” (Ministry of Information Technology
2005). e-Governance, or computerised services, was a second-
ary goal. In the first phase of the project, centres were to focus
on providing “e-literacy” to at least one member of every
household in the state while e-Governance was to be incorpo-
rated as a service offering in the second phase, which began
more than a year after the first centres were opened.
Despite the fact that computerised services were not the
primary goal, it was expected that this would be a core
element of the initiative as it evolved. By 2008, however,
Akshaya centres had virtually no public services in its net-
work of more than 1,000 centres, and all of these were serv-
ices available in FRIENDS. Many other services demanded by
rural citizens remained unavailable, limiting the relevance of
centres to citizen needs and reducing potential income for the
Akshaya centre owners.
Key to understanding the implementation of the Akshaya
initiative is the nature of the ruling coalition when it was
launched. The UDF, led by Congress, was primarily supported
83
SPECIAL ARTICLE
by the Muslim League with 16 seats and the Kerala Congress(M)
with nine seats. The lack of public services in Akshaya centres
is viably linked to the politics of this coalition in two ways.
First, the important Kerala Congress(M) controlled both the
revenue department and rural development, which together
account for a large number of citizen-facing services, but none
of these were provided through Akshaya.
Second, a number of services provided by FRIENDS were not
provided in the Akshaya centres, such as payments for water
bills, property taxes, driver’s licences, and vehicle registration.
Other coalition parties controlled the departments overseeing
these services, with the Kerala Congress (Jacob) and the Ker-
ala Congress(B) overseeing water and transport, respectively,
while Kerala Congress(M) oversaw revenue, which controls
property taxes. Even though the basic infrastructure was in
place to extend these services through Akshaya outlets, this
was not done under the Congress administration. Instead, citi-
zens must either go to the district FRIENDS office or avail of
these services through traditional department-based offices.
Coalition politics, then, played an important role in shaping
technology-enabled service provision in Kerala. While the
state can be seen as an innovator in service reforms, at least in
terms of early service centre implementation relative to other
states, the characteristics of the centres themselves reveal the
power, and interest, of supporting coalition members to retain
direct control over the services in their domain.
Uttar Pradesh
Uttar Pradesh provides a final case for evaluating the effects of
both petty corruption and coalition politics on computerised
service centres. During the period in which centres were initi-
ated, the Samajwadi Party (SP) ruled the state in coalition gov-
ernment. This government supported the expansion of a dis-
trict level rural service centre initiative and also launched
computerised centres in urban areas. However, in both cases
the number of services offered is highly limited and, particu-
larly in the urban case, the implementation of services within
centres led to limited benefits for citizens.
A district collector initiated the rural “Lokvani” project and
was subsequently moved into the department of information
technology in order to lead the extension of the initiative into
other districts. The main goal of Lokvani was to offer an im-
proved platform for citizens to submit grievances to the govern-
ment. Additional services were those that were relatively easy to
offer once the centres were online, such as information about
schemes, which was already published online by relevant de-
partments.21 The other services made available are those most
commonly offered by the district collector, such as certificates.
Additional services overseen by the rural development depart-
ment, which was overseen by a minister from the supporting
Congress Party, were not made available through Lokvani.
A second service centre initiative, e-Suvidha, was later
launched to serve urban areas. In this case, it is difficult to
evaluate the effects of coalition politics, because the SP re-
tained control over nearly all of the departments providing
services directly to citizens. Food and civil supplies was the
84
only department overseeing a service frequently provided in
service centres, ration cards, that a minister from the SP did
not hold, and this service was not made available in e-Suvidha.
The independent MLA holding this position was seen as a key
supporter of the lead party but he also had a criminal record
(he was released from prison on terrorism charges when the
government came to power) and it is perhaps not surprising
that the opportunity to increase government transparency in
this department has not translated into any tangible results.
Within the SP itself, or its successor the Bahujan Samaj
Party, there also seems to have been little interest in providing
comprehensive computerised services in urban areas. The
e-Suvidha initiative was initially met with great optimism,
with one media report claiming that, “People can now look
forward to a redefined concept of public service that obviates
tediousness of long queues moving from one department to
another to procure routine departmental information or pay
public utility bills” (“Now, e-Suvidha to Redefine Public Service”,
Times of India, 11 February 2005), but this was not to be.
Only a short time after the urban centres were launched,
reports began to emerge about poor quality service delivery.
Citizens still have to jostle for space to pay their bills, apart from hav-
ing to undergo an endless wait while in queue. And more often than not,
some arrive at the centres only to find out that some of the facilities are
not available, as claimed otherwise by the government (“E-Suvidha:
Residents Resent Snag-Fraught Facility”, Indian Express, 26 February 2008).
As one citizen lamented,
I don’t understand how the initiative has been serving its purpose. The
queues are still long, and most of the facilities that the centre claims to
offer are not available here (Kumar quoted in “E-Suvidha: Residents
Resent Snag-Fraught Facility”, Indian Express, 26 February 2008).
Given problems with additional services provided in the
e-Suvidha centres, including railway tickets and mobile phone
payments, the future of the initiative is highly uncertain (Times
of India 2009). Though e-Suvidha was launched with great hope
for improving service delivery in urban areas, the state govern-
ment has failed to provide a range of high quality, computerised
services. The combination of high levels of petty corruption and
the need to maintain a delicate coalition balance resulted in
service centres offering little more to citizens than a new loca-
tion for inefficient, and most likely corrupt, service delivery.
Conclusions
This discussion has highlighted the risks associated with pro-
moting technology-based reforms without due attention to the
threats and benefits these policies may present to established in-
terests in the political system. If politicians depend on access to
corrupt income to support their re-election campaigns, then we
should expect them to resist any policies that threaten this eco-
nomic resource. Even where politicians promote reforms, they
are likely to do so in a way that primarily benefits their preferred
constituents. Efforts to improve the quality of service delivery to
all citizens must take the broader causes of problems in service
delivery, and in particular the underlying causes of demands for
bribes, into consideration in order to achieve significant and
long-term improvements in the quality of public services.
June 23, 2012 vol xlvII no 25 EPW Economic & Political Weekly

Notes
1 While the analysis continued after 2006, when
the NeGP was launched, I consider only those
state government initiatives not originating
with the central government programme.
2 There are multiple reasons a citizen might pay
a bribe to receive a service, including access to
a public benefit to which they are not entitled.
However, many bribes are paid simply to re-
ceive basic services in a timely manner (Trans-
parency International India and CMS 2005)
and more efficient service delivery should be
feasible with computerised procedures.
3 Data collected by the author.
4 Interview. Rajasthan State Administrative,
Officer, 9 May 2007.
5 Interview. Indian Administrative Officer,
Chhattisgarh, 23 January 2009.
6 In 2001, the Indian central government pre-
empted state policies in the north-east and
Jammu and Kashmir by implementing a cen-
trally designed service centre initiative in these
states, the Citizen Information Centres (CICs).
As a result, I was not able to analyse the factors
influencing sub-national policy characteristics
in these states.
7 In all analyses, I use the measure of petty cor-
ruption discussed above.
8 “High Corruption Potential” and “High Bribe”
services were determined based on data in
Transparency International India and CMS
(2005).
9 Author’s data collection and analysis. Some
states do not add to 100% because services not
targeted to socio-economic groups are excluded.
10 Interview, Chhattisgarh IT Society Employee,
23 January 2009.
11 Interview, Chhattisgarh IT Department Offi-
cial, 23 January 2009.
12 Policy documents released during the Con-
gress administration allow for a comparison of
the initially offered services with those intro-
duced after the BJP came to power.
13 Interview, Chhattisgarh Information Technol-
ogy Society Representative, 23 January 2009.
14 Interview, Consultant to Tamil Nadu govern-
ment, 17 May 2007.
15 Interview, ELCOT representative, 16 February
2006.
16 Interview, Tamil Nadu IT Department official,
16 February 2006
17 Interview, Tamil Nadu IT Department official,
18 May 2007.
18 Interview, Consultant to Tamil Nadu Govern-
ment, 17 May 2007.
19 While the Congress and the Communist Party
of India (Marxist) are constant leaders of the
two major coalitions, the parties that support
them in these coalitions are more variable
(Nirmala 2006). Because supporting parties
can switch coalitions and because rule of the
state frequently swings from one coalition to
the other, Congress and CPI(M) leaders must
be conscious of retaining the support of smaller
parties in order to stay in power.
20 Both surveys (Transparency International India
and CMS 2005) and anecdotal reports find that
bill payments rarely involve major demands for
bribes. This is likely because a bureaucrat in-
volved in this type of transaction cannot delay
in giving a document or other service to the citi-
zen; they are simply expected to take the citi-
zen’s money for services already delivered.
21 Interview, Uttar Pradesh IT Department Offi-
cial, 13 February 2008.
References
ADPIP (2003): “Friends, Thiruvananthapuram,
Kerala”, Case Study.
Economic & Political Weekly EPW June 23, 2012 vol xlvII no 25
Ahuja, M and A P Singh (2006): “Evaluation of
Computerisation of Land Records in Karnataka”,
Economic & Political Weekly, 7 January, 69-77.
Banerjee, Abhijit and Lakshmi Iyer (2005): “History,
Institutions, and Economic Performance”,
American Economic Review, 95: 1190-1213.
Bardhan, Pranab (2005): “An Economist’s Approach
to the Problem of Corruption”, World Develop-
ment, 34: 341-48.
– (1997): “Corruption and Development: A Review
of Issues”, Journal of Economic Literature, 35:
1320-46.
Bhatnagar, Subhash and Nupur Singh (2009):
“Results from a Study of Impact of E-govern-
ment Projects in India”, Proceedings of the
Third International Conference on ICTD.
Bussell, Jennifer (2010): “Why Get Technical? Cor-
ruption and the Politics of Public Service Re-
form in the Indian States”, Comparative Politi-
cal Studies, 43: 1230-57.
Caseley, Jonathan (2004): “Public Sector Reform and
Corruption: CARD Façade in Andhra Pradesh”,
Economic & Political Weekly, 13 March, 1151-56.
Davis, Jennifer (2004): “Corruption in Public Serv-
ice Delivery: Experience from South Asia’s Wa-
ter and Sanitation Sector”, World Development,
32: 53-71.
de Zwart, Frank (1994): The Bureaucratic Merry-go-
round: Manipulating the Transfer of Indian Civil
Servants (Amsterdam: Amsterdam University
Press).
Election Commission of India (2007): Handbook for
Candidates, Electoral Commission of India,
New Delhi.
Financial Express (2010): “Beyond Corruption in Ma-
harashtra”, 11 November, http://www.financial-
express.com/news/beyond-corruption-in-ma-
harashtra/711067 /0
Frankel, Francine (2005): India’s Political Economy:
1997-2004, second edition (New Delhi: Oxford
University Press).
Heller, Patrick (2000): “Degrees of Democracy:
Some Comparative Lessons from India”, World
Politics, 52: 484-519.
Government of India (2006a): National e-Governance
Plan, Ministry of Information Technology, viewed
on 12 December 2010.(http://mit.gov.in/content/
national-e-governance-plan).
– (2006b): India: e-Readiness Assessment Report
2006: For States/Union Territories, Depart-
ment of Information Technology, New Delhi.
Iype, George (2004): “How Much Does an Election
Cost?”, Rediff Election Special, 25 March,
viewed on 14 December 2010. (http://www.
rediff. com/election/2004/mar/26espec.htm)
Jenkins, Rob (2006): “Democracy, Development
and India’s Struggle against Corruption”, Pub-
lic Policy Research, September.
Kiran, G R (2002): “Front-End First: Citizen Pay-
ment at FRIENDS Centres in Kerala”, UNPAN
eGovernment for Development Success/Fail-
ure Case Studies, #17, viewed on 14 December
2010 (http://www.egov4dev.org/friends.htm).
Kumar, G G (2004): “Kerala: Verdict against Non-
Performance and Congress Factionalims”, Eco-
nomic & Political Weekly, 18 December, 5498-5501.
Kuriyan, Renee and Isha Ray (2009): “Outsourcing
the State? Public-Private Partnerships and In-
formation Technologies in India”, World Deve-
lopment, 37: 1663-73.
Madon, Shiron and G R Kiran (2002): Information
Technology for Citizen-Government Interface:
A Study of FRIENDS Project in Kerala, World
Bank Global Knowledge Sharing Programme.
Manikandan, K (2008): “Innovative Rural e-Gov-
ernance Project Plagued by Neglect”, The Hin-
du, 21 January, viewed on 14 December 2010.
(http://www.thehindu.com/2008/01/21/sto-
ries/2008 012154300600.htm)
Mathew, R (1998): “State Information Technology
Policy – A Critique”, lecture, 26 July, viewed
SPECIAL ARTICLE
on 14 December 2010 (http://expert-eyes.org/
archive/ itcritiq.html).
Mayaram, Baba (2004): “Chhattisgarh: BJP Con-
solidates Its Hold”, Economic & Political Weekly,
18 December, 5516-18.
Ministry of Information Technology (2005): “Akshaya
– An IT Dissemination Project”, http://www.
mit.gov.in/sites/upload_files/dit/files/down-
loads/eready2005/Chap4.pdf
NDTV (2010a): “Alleged Commonwealth Games
Corruption”, 10 August, viewed on 12 December
2010 (http://www.ndtv.com/article/india/al-
leged-commonwealth-games-corruption-kalma-
di-defends-himself-in-a-letter-to-mps-43287).
– (2010b): “2G Spectrum Scam: A Raja Meets Karu-
nanidhi”, 11 December, viewed on 12 December
2010 (http://www.ndtv.com/article/india/2g-
spectrum-scam-a-raja-meets-karunani-
dhi-71983).
Nirmala, P S (2006): “Assembly Polls 2006: A For-
gone Outcome in Kerala?”, Economic & Political
Weekly, 11 March, 866-68.
Oldenburg, Philip (1987): “Middlemen in Third-
World Corruption”, World Politics, 39, 508-35.
Pandey, S (2010): “The Hell Diggers”, Tehelka,
3 April.
Prasad, G K and CSDS Team (2004): “Alliance Ef-
fect, Swing Factor Propelled DPA Victory”, The
Hindu, 20 May, viewed on 14 December 2010
(http://www.hinduonnet.com/elections2004/
verdict2004/stories/2004052000180300.htm).
Rekhi, S and G C Shekhar (1996): “Funding the
Politician”, India Today, 21: 28-40.
Sharma, Ravi (2008): “Gaining Currency”, Frontline,
7-20 June.
Sreekumar, T T (2002): “Civil Society and State-led
Initiatives in ICTs: The Case of Kerala”, Infor-
mation Technology in Developing Countries, De-
cember, viewed on 14 December 2010 (http://
www.iimahd.ernet.in/egov/ifip/dec2002/arti-
cle2.htm).
Srinivasan, Janaki (2005): “Sustainable Access in
Rural India (SARI) Project, Madurai, India” in
Information and Communication Technologies
for Development: A Comparative Analysis of Im-
pacts and Costs from India, IIIT, Bangalore.
Thirunavukkarasu, R (2001): “Tamil Nadu Elec-
tions 2001: Changing Equations”, Economic &
Political Weekly, 7 July, 2486-89.
Toyama, Kentaro, Karishma Kiri, M L Ratan,
A Nileshwar, Rama Vedashree and R F MacGregor
(2004): “Rural Kiosks in India”, Microsoft
Corporation.
Transparency International and CMS (2005): “India
Corruption Study 2005”, http://www.transpar-
encyindia.org/resource/survey_study/
India%20Corruption%20Study%202005.pdf
Uppal, Yogesh (2009): “The Disadvantaged Incum-
bents: Estimating Incumbency Effects in Indi-
an State Legislatures”, Public Choice, 138: 9-27.
Vithal, B P R (1997): “Evolving Trends in the Bureau-
cracy” in P Chatterjee (ed.), State and Politics in
India (New Delhi: Oxford University Press).
Wade, Robert (1985): “The Market for Public Office:
Why the Indian State Is Not Better at Develop-
ment”, World Development, 13: 467-97.
available at
Delhi Magazine Distributors
Pvt Ltd
110, Bangla Sahib Marg
New Delhi 110 001
Ph: 41561062/63
85